The invention relates generally to the production of polyethylene resins, and particularly, to the production of polyethylene resins for use in blow molding applications.
Polyethylene has been used in the production of various blow molded products, such as bottles. Examples of such products include household industrial containers, such as bleach bottles, detergent bottles and the like. Blow molding is accomplished by extruding molten polyethylene resin as a parison or hollow tube into a mold cavity while simultaneously forcing air into the parison so that the parison expands, taking on the shape of the mold. The molten polyethylene cools within the mold until it solidifies to produce the desired molded product.
During blow molding, the polyethylene will typically expand or swell upon exiting the die of the extruder. Although some swelling of the extruded resin may be acceptable and even desirable, excessive swelling can result in defective products and production difficulties. Excessive parison swell can result in molded parts having flash in non-pinchoff areas and/or neck/tail flash which is difficult to trim. In general, more swell makes it more difficult to control the wall thickness of the parison for optimal wall thickness distribution of the products made. Too little swell can be detrimental as well, making it particularly important to control the degree of swell. When a resin exhibits too little swell, it can become difficult or impossible to properly fill out the extremities of the mold, such as the handle of a blow molded bottle. It is, therefore, desirable to control swelling in blow molding operations to within acceptable limits.
U.S. Pat. No. 5,486,575 discloses improving certain properties of chromium produced polyethylene blow molding resins through the addition of organic peroxides. The improvements were directed towards increasing the stress crack resistance of blow molded polyethylene bottles. The improved results were seen in polyethylene resins having a broad molecular weight distribution, i.e. Mw/Mn being greater than 10. Although acceptable swell characteristics were achieved with respect to those polyethylene resins having the broad molecular weight distributions described, poor swell characteristics were observed for polyethylenes having more narrow molecular weight distributions.
Polyethylene resins produced with Ziegler-Natta catalysts have also been modified through the use of free radical initiators, such as oxygen, to improve the polymer properties. U.S. Pat. Nos. 5,728,335 and 5,962,598 discuss the modification of polyethylene resins used in films through such means. Ziegler-Natta produced polyethylene resins used in blow molding resins are typically bimodal resins wherein a low molecular weight polymer and a high molecular weight polymer are combined to provide a broad molecular weight distribution to improve the melt properties of the resin.
It is desirable to improve the swell characteristics of polyethylene blow molded resins having more narrow molecular weight distributions than has heretofore been achieved. It is also desirable to improve other properties of polyethylene resins and to improve upon the processability of such polyethylene resins, particularly those used in connection with blow molded products.
A method of processing polyethylene, and the polyethylene produced by such method, for use as an extruded blow molding resin to improve the properties and processability thereof is provided. The method comprises modifying a polyethylene resin having a molecular weight distribution (Mw/Mn) of less than about 10 by the addition of a free radical initiator to the resin during extrusion of the resin to increase the rheological breadth of the polyethylene from about 10 to about 60% compared to unmodified polyethylene resin extruded under similar conditions.
In certain embodiments, the polyethylene resin may have a melt flow index of at least 1 or greater prior to extrusion. The modified polyethylene produced by such method may also have a YI value reduced by at least about 10% compared to unmodified polyethylene resin extruded under similar conditions.
The free radical initiator may be any source capable of generating a free radical. Examples include air, oxygen or peroxide, which may be added in the amount of from 5 to 100 ppm. The polyethylene resin may be a chromium-catalyst produced polyethylene. The polyethylene resin may also include polyethylene and ethylene copolymers of C3 to C10 alpha-olefins.
In another embodiment of the invention, a method of processing polyethylene resin having desired properties for use as a blow molding resin is provided. The method comprises selecting a first unextruded polyethylene resin having a first initial melt index that provides a desired final melt index after extrusion of the selected polyethylene resin. The first polyethylene resin is modified during extrusion with a selected amount of a free radical initiator so that the first polyethylene resin is modified to provide a modified melt index of the first polyethylene resin that is different from the desired final melt index after extrusion. A second unextruded polyethylene resin is selected having a desired initial melt index based upon the differences between the modified melt index and the desired final melt index to provide a second desired final melt index upon modification of the second polyethylene resin with the selected amount of the free radical initiator after extrusion. This second polyethylene is modified with the free radical initiator while the second polyethylene resin is in a molten state and extruded.
In certain embodiments, the first and second desired final melt indices may be substantially equal, and may be modified with the selected amount of free radical initiator. The desired initial melt index may be greater than the initial melt index of the first polyethylene resin, and the modified second polyethylene resin may have a greater rheological breadth compared to the first polyethylene resin upon extrusion.
Additionally, the unmodified first polyethylene resin and the modified second polyethylene resin may have final melt indices after extrusion that are substantially the same, and wherein extruder pressures for the modified polyethylene resin are less than those for the unmodified polyethylene resin during extrusion.
In another aspect of the invention, a method of processing polyethylene for use as an extruded blow molding resin to improve the color properties thereof is provided. The method comprises modifying a polyethylene resin by the addition of a free radical initiator to the resin during extrusion of the resin to obtain a polyethylene resin having a YI value reduced by at least about 10% compared to unmodified polyethylene resin extruded under similar conditions.
It has been found that by the addition of long-chain branching of polyethylene resins for use in blow molding applications by the utilization of a free radical initiator with the polyethylene prior to extrusion, improvements in the properties and processability of the resins can be achieved. The invention has particular application to those polyethylene resins having narrow molecular weight distributions. Improvements in the way of increased rheological breadth of the ethylene polymers are achieved. In particular, the rheological breadth of the polyethylene can be increased from about 10% to about 60% compared to unmodified polyethylene resin extruded under similar conditions. This directly bears on and results in improvements in the processability, as well as improvements in swell characteristics and other properties of the blow molded resin. Improvements in the color of the extruded polyethylene resin have also been observed by the addition of the free radical initiator.
The polyethylene used for blow molding applications may be polyethylene homopolymers or copolymers of ethylene and C3-C10 alpha-olefin monomers. When copolymers are used, the ethylene content will typically range from about 90 to about 100 mol %, with the balance being made up of the C3-C10 alpha olefin. The method of the invention has particular application in increasing the rheological breadth of polyethylene resins having a narrow molecular weight distribution MWD, which can be used in blow molding applications. Narrower MWD polyethylenes are typically used for film, sheet, fiber and injection molding applications where melt strength and processability is sacrificed for certain solid state properties and aesthetics. Broad MWD distribution resins are typically used in applications where melt strength, processability and solid state performance such as stress crack resistance are desired such as blow molding and pipe applications. The MWD can be described by means of the curve obtained by gel permeation chromatography. However, the MWD is generally described by what is known as the polydispersity index (D), which represents the ratio between the weight-average molecular weight and the number-average molecular weight (Mw/Mn), as defined by standard and well known GPC or SEC methods. The polyethylene resins used have a fluff melt index MI2 (MI using 2.16 kg -ASTM D1238) of at least 1.0 or greater, in particular, the present invention has application to those resins having an MI2 ranging from about 0.75 to about 2, more particularly from about 1.0 to about 1.3.
In the present invention, polyethylene having a molecular weight distribution (Mw/Mn) of less than about 10 were treated to improve the rheological breadth of polyethylene for use in blow molding applications. In particular, the method has application for the processing of polyethylene resins having a molecular weight distribution Mw/Mn of from about 2 to about 9, and more particularly a Mw/Mn from about 5 to about 8. As used herein, Theological breadth refers to breadth of the transition region between Newtonian and power-law type shear rate dependence of the viscosity. The rheological breadth is a function of the relaxation time distribution of the resin, which in turn is a function of the resin""s molecular architecture. It is experimentally determined assuming Cox-Merz rule by fitting flow curves generated using linear-viscoelastic dynamic oscillatory frequency sweep experiments with a modified Carreau-Yasuda (CY) model,                               η          =                                                    η                0                            ⁡                              [                                  1                  +                                                            (                                              λ                        ⁢                                                  xe2x80x83                                                ⁢                                                  γ                          .                                                                    )                                        a                                                  ]                                                                    n                -                1                            a                                      ⁢                  xe2x80x83                                    (        1        )            
where
xcex7=viscosity (Pa s)
xcex3=shear rate (1/s)
a=rheological breadth parameter [CY model parameter which describes the breadth of the transition region between Newtonian and power law behavior]
xcex=relaxation time sec [CY model parameter which describes the location in time of the transition region]
xcex70=zero shear viscosity (Pa s) [CY model parameter which defines the Newtonian plateau]
n=power law constant [CY model parameter which defines the final slope of the high shear rate region]
To facilitate model fitting, the power law constant (n) is held to a constant value (n=0.2). Experiments were carried out using a parallel plate geometry and strains within the linear viscoelastic regime over a frequency range of 0.1 to 316.2 secxe2x88x921. Frequency sweeps were performed at three temperatures (170xc2x0 C., 200xc2x0 C. and 230xc2x0 C.) and the data was shifted to form a mastercurve at 190xc2x0 C. using known time-temperature superposition methods.
For resins with no differences in levels of long chain branching (LCB), it has been observed that the rheological breadth parameter (a) is inversely proportional to the breadth of the molecular weight distribution. Similarly, for samples which have no differences in the molecular weight distribution, the breadth parameter (a) has been found to be inversely proportional to the level of long chain branching. An increase in the rheological breadth of a resin is therefore seen as a decrease in the breadth parameter (a) value for that resin. This correlation is a consequence of the changes in the relaxation time distribution accompanying those changes in molecular architecture.
The present invention has application to those resins having moderate levels of long chain branching resulting from the polymerization process. The level of long chain branching is quantified in terms of the resins flow activation energy (Ea). The time dependent shifts (e.g., horizontal shift of modulus or stress versus frequency) required to form a mastercurve from the flow curves at 170 xc2x0C., 200xc2x0 C. and 230xc2x0 C. are used to calculate the flow activation energy using the well known temperature dependence of the linear viscoelastic properties in the form of the Arrhenius equation,                               α          T                =                  exp          (                                                    E                a                            R                        ⁢                          (                                                1                                      273                    +                    T                                                  -                                  1                                      273                    +                                          T                      o                                                                                  )                                                          (        2        )            
where
Ea=flow activation energy (kcal/mol)
T=temperature of the data being shifted
T0=reference temperature
R=gas constant
xcex1T=shift factor required to superimpose the flow curves at each temperature to the reference temperature (T0)
The flow activation energy is solved using the values of the shift factor required to overlap the flow curve at temperature (T) to that of the flow curve at temperature (T0).
The flow activation energy (Ea) represents the activation energy barrier associated with the energy required to create a hole big enough for a molecule to translate into during flow. This general definition of (Ea) suggests its relationship or sensitivity to changes in molecular architecture such as those associated with changes in levels or types of long chain branching. The polyethylene used in the present invention may be that made using a chromium catalyst that has a fluff flow activation energy (Ea) in the range of 7.25+/xe2x88x920.25 Kcal/mol, which represents a significant amount of long chain branching. A more linear polyethylene made using Ziegler-Natta type catalysts having a similar polydispersity has very low levels of long chain branching such that the fluff flow activation energy of 6.5+/xe2x88x920.25.
The polyethylene is preferably that produced from chromium catalysts capable of producing the narrow molecular weight distribution polyethylene discussed above. The chromium catalysts that are used are those that are well known to those skilled in the art. Activated chromium catalysts on a silica or titanium oxide support are particularly well suited to the polymerization of ethylene for blow molding resins. Increased rheological breadth of polyethylene produced from other catalysts used in the polymerization of olefins, such as Ziegler-Natta, metallocene or late-transition metal catalysts can be obtained as well.
In the method of the invention, a free radical initiator is added to the polyethylene resin prior to extrusion. The free radical initiator, as used herein, is that which results in light crosslinking or branching of the polyethylene molecules. Such free radical initiators include peroxides, oxygen, air and azides. Radiation may also be used as the free radical initiator. The preferred free radical initiators, however, are peroxides, particularly the organic peroxides. Two classes of organic peroxides have been found to be particularly suitable. These are the dialkyl and peroxyketal type peroxides. An example of a commercially available dialkyl peroxide for use as a free radical initiator is 2,5-dimethyl-2,5-di(t-butylperoxy)hexane, available as LUPERSOL 101, from Atofina Chemicals, Inc. Two commercially available peroxyketal peroxides are LUPERSOL 233 and 533, which are examples of t-butyl and t-amyl type peroxides, respectively, and are also available from Atofina Chemicals, Inc.
The choice of peroxide may vary, however, depending upon the particular application and extruder temperatures encountered. Typical extruder temperatures are from about 350xc2x0 F. to about 550xc2x0 F. It is important that the extruder temperature or polyethylene melt be above the decomposition temperature of the peroxide. Thus, extruder temperatures will typically be at least 5% or higher than the decomposition temperature of the peroxide being used to ensure complete decomposition. The extruder temperature can be determined using a combination of peroxide half life versus temperature data and the residence time in the extruder as prescribed by the desired throughput.
The peroxide can be added to the polyethylene fluff or powder prior to introduction into the extruder. For polyethylene fluff having a MI2 of 1.0 or greater, it is usually preferable to add the peroxide to the fluff prior to extrusion. In such cases, the peroxide should be thoroughly mixed or dispersed throughout the polymer before being introduced into the extruder. Alternatively, the peroxide can be injected into the polyethylene melt within the extruder. The peroxide is usually added as a liquid, although the peroxide may be added in other forms as well, such as a peroxide coated solid delivery. The peroxide may also be added or combined with the polyethylene prior to or after the polyethylene is fed into the extruder. It is preferable to add liquid peroxide to the melt phase of the polyethylene within the extruder to ensure that the peroxide is completely dispersed. The peroxide may be introduced into the extruder through any means known to those skilled in the art, such as by means of a gear pump or other delivery device. If oxygen or air is used as the initiator, these are preferable injected into the extruder within the polyethylene melt.
The amount of peroxide or initiator necessary to achieve the desired properties and processability may vary. The amount of peroxide or initiator is important, however, in that too little will not achieve the desired effect, while too much may result in undesirable products being produced. Typically, for peroxides, the amounts used are from about 5 to about 100 ppm, with from about 5 to 50 ppm being more typical. A preferred range for peroxide is that from about 5 to about 40 ppm
In testing, as illustrated by the examples which follow, improved or increased rheological breadth and improved swell for polyethylene resins having narrow molecular weight distributions were obtained for given polyethylene resins. This was accomplished through the addition of a peroxide as the free radical initiator in appropriate amounts to achieve the increased rheological breadth. Improvements in other characteristics, such as color, were also observed.